A systematic review and meta-analysis of randomized controlled trials (RCTs) were performed to understand the effectiveness of medication adherence (MA) interventions among Chinese patients with hypertension.

Methods

A literature search was conducted with three English databases (PubMed, Web of Science and Embase) and three Chinese databases (China National Knowledge Infrastructure, Wanfang and VIP Database for Chinese Technical Periodicals) for the period from 1970 to October 2017. Only both RCTs with a minimum of 10 participants in each intervention group and Chinese patients with hypertension as participants were included. A random-effects model was applied to calculate pooled effect sizes with 95% CI. Subgroup analysis was conducted to identify potential sources of heterogeneity from duration of intervention, type of intervener, methods of intervention and sites of intervention. Funnel plots and Egger's test were used to evaluate for publication bias.

Key findings

A total of 48 studies met criteria for the meta-analysis, including 14 568 participants, testing 57 independent comparisons. Overall, the effect size revealed that interventions significantly improved MA (pooled relative risk = 1.59, 95% CI: 1.43 to 1.78; pooled Cohen's d = 1.42, 95% CI: 0.976 to 1.876). Interventions were found to significantly reduce blood pressure (BP) (systolic BP: Cohen's d = −0.85, 95% CI: −1.11 to −0.60 and diastolic BP: Cohen's d = −0.73, 95% CI: −1.00 to −0.46). Longer duration of intervention gave better effectiveness. Physician as interventionist, regular follow-up visits and interventions conducted at a hospital were associated with better effectiveness.

Conclusion

Adherence interventions improve MA and reduce uncontrolled BP among Chinese patients with hypertension. In the future, investigators should adopt a skill set to address the problem of poor MA.

Introduction

Hypertension is a critical public health problem worldwide and is considered the most dangerous factor for cardiovascular disease, cerebrovascular disease and renal diseases. In China, prevalence of hypertension has been rising, with rates increasing from 18.8% in 2002 to 25.2% in 2012.1 The health disparity is particularly on hypertension control. The Chinese population has a lower prevalence of controlled hypertension (15%) than does the American population (55.7%).2, 3 Hypertensive patients’ blood pressure (BP) needs be controlled, to decrease the incidence of adverse sequelae. There is an association between poor medication adherence (MA) and inadequate BP control. Nearly 75% of non-adherent patients fail to achieve optimal BP.4 Adherence to prescribed antihypertensive agents can reduce the risk of stroke by 35–40% and myocardial infarction by 20–25% according to the survey by Chobanian et al.5 and Collins et al.6 However, non-adherence to antihypertensive therapy is increasingly prevalent, causing approximately 125 000 deaths per year.7 Factor analysis reveals that poor MA may be related to the symptomless nature of the condition, the long duration of therapy, adverse drug reactions, complicated drug regimens, the lack of understanding about hypertension management, expensive drugs and the challenges to individual patients’ health beliefs.8, 9 Some authorities point out that 80% adherence is probably sufficient for most drugs.10, 11 Chinese hypertensive patients display very poor MA, with reported rates ranging from 40% to 65.8%.12-15 Due to the large population base and the rapid increase in better living conditions in China, interventions to improve patients’ MA are more important now than ever. In recent years, using intervention measures and experience of developed countries for reference, Chinese researchers have conducted some interventions to improve patients’ MA and clinical outcomes. However, conclusions from epidemiological studies about the intervention effectiveness differ widely.

Therefore, a systematic review and meta-analysis were conducted to summarize evidence for effectiveness of MA interventions among Chinese patients with hypertension based on available RCTs. Heterogeneity and publication bias were quantified and evaluated independently.

Methods

Standardized systematic review and meta-analysis were adopted to synthesize the effectiveness for interventions to improve the MA among Chinese patients with hypertension. Findings are reported following the PRISMA guidelines.

Inclusion criteria

The inclusion criteria were as follows (PICOs): (1) population: Chinese adults, male or female, aged more than 18 years with a diagnosis of hypertension in each intervention group; (2) intervention: any intervention targeted at MA, with a minimum of 10 patients per group; (3) control: inactive control groups or only basic health education; (4) outcome: MA outcome reported; and (5) study design: only RCTs were considered. Exclusion criteria included case reports, editorials, reviews, letters and comments. There were no exclusions based on publication dates. Only papers written in English or Chinese were included.

Information sources and search strategies

Multiple search strategies were performed to reduce the risk of retrieval bias. An information retriever and two investigators conducted literature searches using three English databases (PubMed, Web of Science and Embase) and three Chinese databases (China National Knowledge Infrastructure, Wanfang and VIP Database for Chinese Technical Periodicals) for the period from 1970 to October 2017 (MA behavior has changed a little since 197016, 17). Retrieved titles, abstracts and then full texts were screened in duplicate against the inclusion and exclusion criteria to identify potential papers. To search the English databases, research terms included medication adherence OR patient compliance OR compliant OR compliance OR adherent OR adherence OR noncompliant OR noncompliance OR nonadherence OR nonadherent OR hypertension OR antihypertensive OR high blood pressure OR hypertensive, pharmaceutical services OR pharmaceutical care OR pharmacist education OR counsel OR medication guideline OR intervention OR China OR Chinese. Chinese databases were searched using subject terms medication adherence, compliance and hypertension. Search strategies for each database are included as Appendix S1.

Data extraction

To reduce the risk of bias, two investigators independently extracted data from eligible studies and entered the data into an Excel spreadsheet (Microsoft Excel 2007), which included the following: first author's name, year of publication, location (province), mean age of study population, total sample size, number of patients in both intervention and control groups’ components and duration of intervention in each study, the number of patients with good MA in both intervention and control groups between both baseline and outcome, and mean systolic blood pressure (SBP) and diastolic blood pressure (DBP) values. Any disagreements were resolved by consensus or by the third investigator.

Risk of bias

Multiple search strategies were conducted among English databases and Chinese databases to avoid publication bias. The presence of publication bias was detected using funnel plots visually and Egger's test statistically.18

Two investigators also assessed the quality of eligible articles independently using Jadad scale.19 The assessment standard was based on the following: randomization, double blind, failure to follow-up and cancellation of investigation. Scoring was according to a 0–5 scale: ≤2 scores were regarded as lower quality, and ≥3 scores were regarded as high quality.

Statistical analysis

Stata 11.0 (Stata Corporation, College Station, Texas, USA). and EndNote X7 (Thomson Reuters Corporation, Manhattan, New York, USA) were used to calculate effect size and to manage citation data, respectively. All eligible studies were grouped and described according to whether they reported continuous data (mean ± standard deviation) or dichotomous data (good MA or poor MA). Standardized mean difference effect size (Cohen's d) and relative risk value (RR) were calculated for each comparison of those MA outcomes as continuous data and dichotomous data, respectively.20 To facilitate the merging of data, days, weeks and years are uniformly converted into months. A forest plot was constructed to reflect the effect size and 95% confidence interval, calculating the I-squared and Q statistic to determine the heterogeneity among all eligible studies (I² = 25%, 50%, 75% delivering corresponding information of low, moderate and high levels of heterogeneity, respectively).21 Heterogeneity was anticipated due to its universality in behavioural research.22 Random-effects models were chosen due to the expected heterogeneity that I² was >50%.21

Subgroup analyses were conducted to explore the effect of intervention duration, intervention components, deliverer of intervention and site of intervention.

Results

A total of 7135 potentially relevant studies were identified by the search strategy. After initial screening, 3590 duplicates were excluded. Then, 1679 and 1244 studies were excluded by reading titles and abstracts, respectively. Six hundred and twenty-two potentially eligible studies were assessed for eligibility. Five hundred and seventy-four studies did not meet inclusion criteria. Finally, 57 separate comparisons from 48 eligible studies were identified, including 12 studies written in English and 36 studies in Chinese. Eligible studies included a total of 9227 intervention participants and 8909 control participants with an overall total of 14 568 participants (see PRISMA chart attached as Figure S1).

Study characteristics

Table 1 provides an overview of the 48 included studies.23-70 Nine studies reported two comparisons23, 24, 27, 30, 32, 33, 42, 61, 67 (designated as different follow-up duration and intervention intensity). A total of 14 568 participants from 20 provinces or regions were included. Ten studies were carried out in municipalities (Beijing, Shanghai and Tianjin), two in special administrative region of China (Hong Kong) and 36 in provinces. Three types of professional (pharmacist, physician, nurse) delivered the interventions at three different sites (home, community, hospital). Four different intervention components were identified including medication education, self-monitoring of BP, reminders and regular follow-up visits. More detailed information on intervention characteristics is included in Table S1. All eligible studies were published between 2003 and 2017, and 45 studies were published in 2010 or later. Thirty-nine studies reported the mean age of patients, with the median age of 62.2 years. Additional details are displayed in Table 1.

Table 1. Characteristics of primary studies

Characteristic	n	Min	Q1	Mdn	Q3	Max
Year of publication	48	2003	2013	2014	2016	2017
Mean age (years)	39	46.8	56.8	62.2	67.9	77.4
Number of women	43	19	44	61	109	1304
Number of intervention participants	48	25	45	66	127	1062
Number of control participants	48	25	43	62	129	1150
Duration of intervention (month)	45	1	3	6	12	36
Number of intervention contents	48	1	2	3	5	10

Max, maximum; Mdn, median; Min, minimum; N, number of studies; Q, quartile.

Quantitative synthesis

The results of the pooled analysis are displayed in Table 2. We chose the random-effects model to calculate pooled effect size due to the high heterogeneity among comparisons. In the analysis of dichotomous data, a total of 39 intervention versus control comparisons were calculated to pool the effect size. The RR value was 1.59 (95% CI: 1.43 to 1.78; P < 0.01, Figure 1). We calculated effect size for 27 intervention pre- versus post-comparisons and 27 control pre- and post-comparisons, for which the RR values were 1.80 (95% CI: 1.46 to 2.21; P < 0.01) and 1.05 (95% CI: 0.902 to 1.22; P = 0.54), respectively. In the analysis of continuous data, the ES (Cohen's d) across the 18 intervention versus control comparisons was 1.42 (95% CI: 0.98 to 1.88; P < 0.01, Figure 2). For 15 intervention group and 14 control group pre- versus post-comparisons, the Cohen's d was 1.57 (95% CI: 0.97 to 2.16; P < 0.01) and 0.231 (95% CI: −0.14 to 0.603; P = 0.47), respectively.

Table 2. Meta-analysis summary statistics and Egger's text

	k	ES (95% CI)	P(ES)	I ²	Q	P(Q)	Egger's
MA, intervention versus control comparisons	39	RR: 1.59 (1.43 to 1.78)	<0.001	93.7	602.44	<0.001	<0.001
MA, intervention pre- versus post-comparisons	27	RR: 1.80 (1.46 to 2.21)	<0.001	97.4	994.93	<0.001	<0.001
MA, control pre- versus post-comparisons	27	RR: 1.05 (0.90 to 1.22)	0.542	88.6	228.4	<0.001	0.101
MA, intervention versus control comparisons	18	d: 1.42 (0.98 to 1.88)	<0.001	95.9	412.72	<0.001	0.001
MA, intervention pre- versus post-comparisons	15	d: 1.57 (0.97 to 2.16)	<0.001	97.2	497.52	<0.001	0.024
MA, control pre- versus post-comparisons	14	d: 0.23 (−0.14 to 0.60)	0.4697	94.5	234.29	<0.001	0.664
SBP, intervention versus control comparisons	31	d: −0.85 (−1.11, −0.60)	<0.001	95.7	703.18	<0.001	0.432
DBP, intervention versus control comparisons	30	d: −0.73 (−1.00, −0.46)	<0.001	96.1	740.39	<0.001	0.788

CI, confidence interval; d, Cohen's d; DBP, diastolic blood pressure; I², heterogeneity index; K, number of comparisons; MA, medication adherence; Q, heterogeneity statistic; RR, relative risk; SBP, systolic blood pressure.

Details are in the caption following the image — **Figure 1**
Open in figure viewer PowerPoint

Forest plot of included two-group studies, intervention versus control groups (dichotomous data). [Colour figure can be viewed at wileyonlinelibrary.com]

The change of BP was also analysed to assess the effectiveness of intervention. A total of 31 comparisons of SBP and 30 comparisons of DBP were included in the analysis. For intervention groups versus control groups of SBP, the overall ES (Cohen's d) was −0.85 (−1.109 to −0.60; P < 0.01) and the overall ES (Cohen's d) of DBP was −0.73 (95% CI: −1.00 to −0.46; P < 0.01).

All types of comparisons showed significant heterogeneity (Table 2). The I² ranged from 88.6% to 97.4%.

Subgroup analysis

The ESs were influenced by the duration of the intervention (P < 0.001). Longer duration gave a larger ES for <6 months versus 6, 7, 8, 9, 10 and 11 months versus 12 months and above (0.843 versus 1.754 versus 3.03) (Table 3). Studies with physician, pharmacist and nurse interventionists reported significantly greater ESs than without physician, pharmacist and nurse interventionists. However, physicians were more effective (2.80) than pharmacist (1.95) or nurse (1.79).

Table 3. Intervention characteristics of moderator analyses

Moderator	k	Cohen's d	95% CI	Q _between	P(Q_between)
Duration of intervention
t < 6 months	10	0.84	0.45 to 1.24	83.21	<0.001
6 months ≦ t < 12 months	5	1.75	0.64 to 2.88
12 months ≦ t	3	3.03	1.32 to 4.80
Interventionist
Pharmacist
Yes	6	1.95	0.94 to 2.95	10.14	0.001
No	12	1.22	0.74 to 1.71	10.14	0.001
Physician
Yes	6	2.80	1.71 to 3.88	136.36	<0.001
No	12	0.80	0.44 to 1.15	136.36	<0.001
Nurse
Yes	6	1.80	0.92 to 2.65	87.56	<0.001
No	12	1.19	0.72 to 1.65	87.56	<0.001
Methods of intervention
Education
Yes	16	1.51	0.98 to 2.04	0.35	0.55
No	2	0.919	0.71 to 1.13	0.35	0.55
Self-monitor of signs
Yes	3	0.58	0.22 to 0.93	5.29	0.02
No	15	1.71	1.13 to 2.29	5.29	0.02
Reminder
Yes	3	0.18	0.02 to 0.34	103.39	<0.001
No	15	1.70	1.18 to 2.22	103.39	<0.001
Regular follow-up visit
Yes	14	1.77	1.18 to 2.36	84.39	<0.001
No	4	0.38	0.03 to 0.73	84.39	<0.001
Place of intervention
At home
Yes	13	1.82	1.21 to 2.44	76.36	<0.001
No	5	0.50	0.15 to 0.86	76.36	<0.001
At community
Yes	6	1.26	0.41 to 2.11	2.24	0.13
No	12	1.51	0.95 to 2.07	2.24	0.13
At hospital
Yes	3	3.79	0.25 to 7.83	1.54	0.21
No	15	1.07	0.65 to 1.19	1.54	0.21

CI, confidence interval; K, number of comparisons; Q_between, heterogeneity statistic.

Adopting education compared to no education had a greater ES (Cohen's d = 1.51 compared to Cohen's d = 0.92) (P = 0.55). Self-monitoring of BP and setting a reminder each gave a lower ES than when these were not done (0.58 versus 1.71, P = 0.02; 0.18 versus 1.70, P < 0.001, respectively). In addition, the difference in ES for interventionists’ regular follow-up visits and absence of follow-up visits was statistically significant (1.77 versus 0.38, P < 0.001).

There was a significant difference when interventions were conducted at home versus not at home (1.82 versus 0.50, P < 0.001). Non-significant variables included interventions delivered in community and interventions delivered in hospital.

Publication bias

Funnel plots were conducted to assess the potential publication bias of four main comparisons, which showed some evidence of publication bias among intervention versus control comparisons (dichotomous and continuous data) with very large positive ESs. The finding was confirmed by Egger's test (P < 0.001, P = 0.001). No evidence of publication bias was showed in funnel plots of BP intervention versus control comparisons, and Egger's test was not significant (SBP: P = 0.432, DBP: P = 0.788). In addition, other comparisons also were detected using Egger's test, which showed higher risk of publication bias among intervention group pre- versus post-comparisons (dichotomous data: P < 0.001, continuous data: P = 0.024) and lower risk of publication bias among control group pre- versus post-comparisons (dichotomous data: P = 0.101, continuous data: P = 0.664). Details are displayed in Table 2 and Figure 3.

Discussion

Findings from this meta-analysis suggest that interventions to improve MA among Chinese patients with hypertension are effective and may lead to less uncontrolled BP. Intervention effectiveness increases with the duration of intervention and regular follow-up visits, as well as the intervention being delivered by a physician.

This review is a comprehensive analysis of MA intervention effectiveness for hypertensive patients in China. Multiple search strategies were conducted using English and Chinese databases to identify as larger studies as possible. However, there are several limitations of this meta-analysis. Although the search strategies were constructed systematically, it is possible that some eligible articles and unpublished articles may have been missed. Dichotomous and continuous outcomes were analysed separately, which might have introduced bias. Interventions of different intensities were included in this analysis, which may increase the risk of publication bias. Some studies did not report relevant results; for example, partial studies have no baseline data of BP. This raises the possibility of information bias. Finally, there is heterogeneity and publication bias across some stratified analyses.

Among 48 eligible studies, including 57 comparisons, 12 control versus intervention comparisons reported that changes had no statistical significance, suggesting disparate outcomes and implying publication bias. The characteristic of intervention of these studies mainly revealed the following points: (1) relevant research started relatively recently, but the attention of researchers grew gradually. (2) These interventions were mainly conducted in economically developed regions, with poorer performance in underdeveloped areas. (3) The content of interventions was more restricted than developed countries.

Of the overall ESs, positive effectiveness was found for interventions to improve MA among Chinese patients with hypertension. This effect size was stronger than that reported by recent foreign meta-analysis71 (which reported multiple racial studies worldwide). Our meta-analysis also suggests that interventions to improve MA among Chinese patients with hypertension can effectively decrease uncontrolled SBP and DBP. However, these results differ from those obtained in a meta-analysis of pharmacist interventions by Machado et al.,72 which reported that only SBP can be affected by intervention, but the studies included in this meta-analysis did not focus on improving only MA. The comprehensive results reveal that adherence to prescribed antihypertensive agents is an important factor (probably more than this one) for achieving optimal BP. The same result also was reported by the systematic review of effectiveness of pharmacist interventions by Morgado et al.73

There were several interesting findings among our subgroup analysis results: larger ES was found with increasing duration of intervention. However, a recent meta-analysis by Ruppar et al.74 on assessing the effectiveness of MA interventions in heart failure patients showed no significant relationship with the length of intervention effectiveness. A possible explanation is that longer intervention duration can change medication habits. In addition, physicians were more effective in improving MA than pharmacists or nurses. This result contrasts with another meta-analysis by Ruppar et al.75 showing that pharmacists play more effective role than others for hypertensive black adults. Clinical pharmacists from hospitals were the majority of all pharmacists as providers of delivery interventions. However, the profession of clinical pharmacist in China has been established for a relatively short time compared with developed countries and is not well recognized by patients.76 Even most medication guidelines and health education were delivered by physician in hospital, which resulted in lower patients’ acceptance to clinical pharmacists’ interventions possibly.

Adopting education to improve the rate of MA was found to have no significantly larger ES than no education. Assuredly, medication education is widely practised and has been shown to be effective.16 However, Chinese educational level is very low, with 31.57% of the population only receiving primary education or below,77 and the situation is worse among patients with hypertension. Therefore, an appropriate approach of medication education needs to be tested to address the issue. The ESs for self-monitoring of signs and setting reminders designed as intervention components were found to be significantly lower than interventions lacking these. Further behavioural research needs to consider the needs of patients of varying levels of education.

Interventions delivered at home were more effective to improve patients’ MA. Interventions delivered at home may reduce the patient's money and time costs, which can improve patient and family receptivity to interventions. The considerable ES for interventions delivered at hospital was larger than that for home or community, but there was no statistical significance, which may be related to not enough number of comparisons.

Conclusion

The meta-analysis showed that interventions can improve MA and reduce uncontrolled BP among Chinese patients with hypertension. Self-monitoring of BP, setting reminders and longer duration of intervention were components of interventions likely to be associated with better MA.

Declarations

Conflict of interest

No author has conflict of interests to declare. All authors had complete access to the study data that support the publication.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (Grant Number 71503006), the General Project of Natural Science of University of Anhui Province (Grant Number KJ 2015A007) and the University Leaders Introduce and Cultivate Programs (grant number gxbjZD2016032)

Authors’ contributions

RX wrote articles, did data extraction and conducted statistical analysis. XX performed article review and language modification. SL conducted data extraction and literature search. XC carried out data extraction and literature search. SW conducted data extraction and statistical analysis. CH carried out statistical analysis and language modification. XL performed article review and language modification.

Ethical approval

This review has not needed to seek ethical approval for the work. This review does not involve animal experiments and volunteer experiments.

Supporting Information

References

1 National Health and Family Planning Commission of China. China Health and Family Planning Statistical Yearbook. Beijing: China union medical university press, 2015.
Google Scholar
2Zhang W et al. A nationwide cross-sectional survey on prevalence, management and pharmacoepidemiology patterns on hypertension in Chinese patients with chronic kidney disease. Sci Rep 2016; 6: 38768.
10.1038/srep38768
CAS PubMed Web of Science® Google Scholar
3Yoon SS, Carroll MD, Fryar CD. Hypertension prevalence and control among adults: United States, 2011–2014. Hyattsville, MD: Department of Health and Human Services, 2015.
Google Scholar
4Burnier M. Drug adherence in hypertension. Pharmacol Res 2017; 125(Pt B): 142–149.
10.1016/j.phrs.2017.08.015
CAS PubMed Web of Science® Google Scholar
5Chobanian AV et al. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 2003; 42: 1206–1252.
10.1161/01.HYP.0000107251.49515.c2
CAS PubMed Web of Science® Google Scholar
6Collins R et al. Blood pressure, stroke, and coronary heart disease. Part 2, Short-term reductions in blood pressure: overview of randomised drug trials in their epidemiological context. Lancet 1990; 335: 827–838.
10.1016/0140-6736(90)90944-Z
CAS PubMed Web of Science® Google Scholar
7Burrell CD, Levy RA. Therapeutic consequences of noncompliance. In: Improving medication compliance. Proceedings of a symposium. Washington, DC: National Pharmaceutical Council. 1984: 7–16.
Google Scholar
8Dowell J, Jones A, Snadden D. Exploring medication use to seek concordance with ‘non-adherent’ patients: a qualitative study. Br J Gen Pract 2002; 52: 24–32.
PubMed Web of Science® Google Scholar
9Ebrahim S. Detection, adherence and control of hypertension for the prevention of stroke: a systematic review. Health Technol Assess 1998; 2: i–iv, 1–78.
CAS PubMed Google Scholar
10Rudd P. Medication packaging: simple solutions to nonadherence problems? Clin Pharmacol Ther 1979; 25: 257–265.
10.1002/cpt1979253257
PubMed Web of Science® Google Scholar
11Dunbar-Jacob J, Mortimer-Stephens MK. Treatment adherence in chronic disease. J Clin Epidemiol 2001; 54(Suppl 1): S57–S60.
10.1016/S0895-4356(01)00457-7
PubMed Web of Science® Google Scholar
12Yi-Bing W et al. Patient related factors for optimal blood pressure control in patients with hypertension. Afr Health Sci 2013; 13: 579–583.
PubMed Web of Science® Google Scholar
13Strand MA et al. A community-based cross-sectional survey of medication utilization among chronic disease patients in China. Int J Pharm Pract 2016; 25: 371–378.
10.1111/ijpp.12327
PubMed Web of Science® Google Scholar
14Cai L et al. Socioeconomic differences in prevalence, awareness, control and self-management of hypertension among four minority ethnic groups, Na Xi, Li Shu, Dai and Jing Po, in rural southwest China. J Hum Hypertens 2017; 31: 388–394.
10.1038/jhh.2016.87
CAS PubMed Web of Science® Google Scholar
15Hou Y et al. The association between self-perceptions of aging and antihypertensive medication adherence in older Chinese adults. Aging Clin Exp Res 2016; 28: 1113–1120.
10.1007/s40520-015-0516-z
PubMed Web of Science® Google Scholar
16Conn VS et al. Interventions to improve medication adherence among older adults: meta-analysis of adherence outcomes among randomized controlled trials. Gerontologist 2009; 49: 447–462.
10.1093/geront/gnp037
PubMed Web of Science® Google Scholar
17Peterson AM, Takiya L, Finley R. Meta-analysis of trials of interventions to improve medication adherence. Am J Health Syst Pharm 2003; 60: 657–665.
10.1093/ajhp/60.7.657
PubMed Web of Science® Google Scholar
18Sterne JA, Egger M. Funnel plots for detecting bias in meta-analysis: guidelines on choice of axis. J Clin Epidemiol 2001; 54: 1046–1055.
10.1016/S0895-4356(01)00377-8
CAS PubMed Web of Science® Google Scholar
19Jadad AR et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17: 1–12.
10.1016/0197-2456(95)00134-4
CAS PubMed Google Scholar
20Borenstein MHL, Higgins JPT, Rothstein H. Introduction to Meta-Analysis. West Sussex: Wiley, 2009.
10.1002/9780470743386
CAS PubMed Google Scholar
21Higgins JP et al. Measuring inconsistency in meta-analyses. BMJ 2003; 327: 557–560.
10.1002/sim.1186
PubMed Google Scholar
22Conn VSHA et al. Metabolic effects of interventions to increase exercise in adults with type 2 diabetes. Diabetologia 2007; 50: 913–921.
10.1007/s00125-007-0625-0
CAS PubMed Web of Science® Google Scholar
23Shen Y et al. Family member-based supervision of patients with hypertension: a cluster randomized trial in rural China. J Hum Hypertens 2017; 31: 29–36.
10.1038/jhh.2016.8
CAS PubMed Web of Science® Google Scholar
24Lu CH et al. Community-based interventions in hypertensive patients: a comparison of three health education strategies. BMC Public Health 2015; 15: 33.
10.1186/s12889-015-1401-6
PubMed Web of Science® Google Scholar
25Yu PL, Ye W, Liu XR. Evaluation on the effectiveness for self-management of hypertensive patients in a community. Zhonghua Liu Xing Bing Xue Za Zhi 2003; 24: 790–793.
PubMed Google Scholar
26Zang XY et al. Effect on blood pressure of a continued nursing intervention using chronotherapeutics for adult Chinese hypertensive patients. J Clin Nurs 2010; 19: 1149–1156.
10.1111/j.1365-2702.2009.03166.x
CAS PubMed Web of Science® Google Scholar
27Peng B et al. Implementation of a structured guideline-based program for the secondary prevention of ischemic stroke in China. Stroke 2014; 45: 515–519.
10.1161/STROKEAHA.113.001424
PubMed Web of Science® Google Scholar
28Jia JJ et al. Impact of physician-coordinated intensive follow-up on long-term medical costs in patients with unstable angina undergoing percutaneous coronary intervention. Acta Cardiol Sin 2017; 33: 173–181.
PubMed Web of Science® Google Scholar
29Ma C et al. Evaluation of the effect of motivational interviewing counselling on hypertension care. Patient Educ Couns 2014; 95: 231–237.
10.1016/j.pec.2014.01.011
PubMed Web of Science® Google Scholar
30Wong MC et al. Effectiveness of a pharmacist-led drug counseling on enhancing antihypertensive adherence and blood pressure control: a randomized controlled trial. J Clin Pharmacol 2013; 53: 753–761.
10.1002/jcph.101
PubMed Web of Science® Google Scholar
31Chiu CW, Wong FK. Effects of 8 weeks sustained follow-up after a nurse consultation on hypertension: a randomised trial. Int J Nurs Stud 2010; 47: 1374–1382.
10.1016/j.ijnurstu.2010.03.018
PubMed Web of Science® Google Scholar
32Huang B et al. Effectiveness of self-management support in maintenance haemodialysis patients with hypertension: a pilot cluster randomized controlled trial. Nephrology (Carlton) 2017; https://doi.org/10.1111/nep.13098. [Epub ahead of print]
10.1111/nep.13098
Google Scholar
33Jiang X, Sit JW, Wong TK. A nurse-led cardiac rehabilitation programme improves health behaviours and cardiac physiological risk parameters: evidence from Chengdu, China. J Clin Nurs 2007; 16: 1886–1897.
10.1111/j.1365-2702.2007.01838.x
CAS PubMed Web of Science® Google Scholar
34Zhu X, Wong FK, Wu LH. Development and evaluation of a nurse-led hypertension management model in a community: a pilot randomized controlled trial. Int J Clin Exp Med 2014; 7: 4369–4377.
PubMed Web of Science® Google Scholar
35Zhong YJ, Li WZ, Liu WH. The clinical pharmacist applies drugs the compliance intervention to the patients with elevated blood pressure. Chin J Clin Rational Drug Use 2010; 13: 6–7.
Google Scholar
36Zhang YZ, Zhang J, Tong l. The efficacy of self-efficacy theory guided nursing intervention on self-administration of medication in elderly patients with hypertension. J Nurs Sci 2015; 30: 74–77.
Google Scholar
37Kang K et al. Influence of pharmacist intervention on medication compliance in aged patients with hypertension. Chin J Cardiovasc Rehabil Med 2012; 21: 635–637.
Google Scholar
38Yin H et al. Analysis of clinical pharmacist's follow-up and medication intervention on patients with hypertension. Clin Med J 2016; 14: 54–56.
Google Scholar
39Lin DQ. Analysis of pharmacist intervention to improve medication adherence among patients with hypertension and type 2 diabetes. Contemporary Med 2014; 20: 104–105.
Google Scholar
40Qian J, Li ZY. Effectiveness of pharmacist intervention to improve medication adherence among patients with hypertension and stroke. Jiangxi Med J 2012; 47: 1089–1091.
Google Scholar
41Sun YP et al. The effect of medication education from clinical pharmacists on the elderly hypertension. Drug Eval 2013; 10: 16–19.
Google Scholar
42Cai WH et al. The effect of the progressive intervention of pharmacist on patients discharged hypertension. Guangdong Med J 2013; 34: 2400–2401.
Google Scholar
43Hao JL, Wang YZ, Li Y. Effect of pharmaceutical care program in peritoneal dialysis patients with hypertension. Anhui Med Pharm J 2014; 18: 970–972.
Google Scholar
44Gu HJ. The effect of pharmacological services on oral blood pressure medication in patients with hypertension. J Med Theor Prac 2017; 30: 612–613.
Google Scholar
45Zhang M et al. Comparison of application of two medication education methods in elderly hospitalized patients with diabetes and hypertension. Pract J Clin Med 2017; 13: 131–133.
Google Scholar
46Xu GX. The role of clinical pharmacists in improving the patient's compliance with medication compliance. Guide China Med 2014; 12: 143.
Google Scholar
47Huang LZ et al. Clinical evaluation of pharmaceutical care on patients with type 2 diabetes mellitus and hypertension. Pharm Today 2017; 27: 42–44.
Google Scholar
48Chen Y et al. Effects of comprehensive geriatric assessment and intervention on medication compliance and blood pressure in elderly patients with hypertension combined anxiety. J Chengde Med College 2017; 34: 111–113.
Google Scholar
49Xia CY, Zhang XH. Effects of the motivational interviewing on medication adherence behaviors in hypertensive patients. China J Mod Med 2012; 22: 66–69.
Google Scholar
50Zhang Y, Wu LY. Observation of the effect of nursing intervention on medication compliance in the aged hypertensive patients. Chin J Pharmacoepidemiol 2014; 23: 305–307.
Google Scholar
51Wang R et al. Effect of transtheoretical model based nursing intervention on medication adherence of patients with hypertension. Nurs J Chin PLA 2015; 32: 31–34.
PubMed Google Scholar
52Wang LJ, Qu WX. Effect of family nursing intervention on home treatment therapeutic compliance of patients with essential hypertension and blood pressure. Med J West Chin 2015; 27: 796–798.
Google Scholar
53Song SY, Dai FY, Wu ZF. Influence of health education on medication compliance in patients with hypertension. Hainan Med J 2017; 28: 2562–2563.
Google Scholar
54Zhang WF. The analysis of health education intervention to improve medication compliance among patients with hypertension. China Pharm 2013; 22: 73–74.
Google Scholar
55Su T, Ji HP. The effect of health education to improve compliance of patients with primary hypertension. Jiangsu J Prev Med 2014; 25: 91–92.
Google Scholar
56Cheng M et al. Effect of continuous health education on medicine-taking compliance and blood pressure of hypertensive patients. BMU J 2016; 39: 364–366.
Google Scholar
57Li HX. The effects of interventionist intervention on the compliance of patients with liver disease combined with hypertension. Cardiovasc Dis J Intergrated Tradi Chinese West Med 2014; 2: 62–64.
Google Scholar
58Wu JJ et al. The effects of cognitive behavioral intervention on the negative emotions and medication compliance of elderly patients with hypertension. J Pract Med 2016; 32: 3264–3267.
Google Scholar
59Zhang ZG et al. Effects of community intervention strategies on treatment compliance of hypertension in rural. Chin J Cardiovasc Med 2014; 19: 108–111.
Google Scholar
60Zhao XJ, Gao JL, Fu H. Influence of community-based group intervention on treatment adherence of hypertensive patients. Chinese General Pract 2011; 14: 1181–1184.
Google Scholar
61Wang S et al. Impact of community nurse-led new team model of community – based chronic disease management on the medication adherence self-efficacy and medication adherence in hypertensive patients. Chinese General Pract 2017; 20: 3343–3348.
Google Scholar
62Liu H et al. The compliance effect of community group service on hypertension management. Shanghai J Prevent Med 2012; 24: 333–335.
Google Scholar
63Zhong Y et al. Clinical effects on patients with hypertension under follow-up intervention. J Med Postgra 2016; 26: 392–394.
Google Scholar
64Qian KK. The effect of WeChat health education platform to influence the compliance of patients with hypertension. North Pharm 2015; 12: 195.
Google Scholar
65Luo JH. The effect of nursing intervention on compliance and hypotensive effect of patients with hospital discharge. Harbin Med J 2015; 35: 411–412.
Google Scholar
66Du J. Study on the influence of pharmacist intervention on medication compliance of elderly patients with hypertension. China Foreign Med Treat 2014; 30: 153–154.
Google Scholar
67Chen XX, He QY, Chen SR. Effect of pharmacist intervention model on compliance and efficacy of treatment in discharged hypertensive patients. Pharm Today 2014; 24: 363–365.
Google Scholar
68Qin L, Wu RR, Liu M. Effect of pharmaceutical care on outpatients with essential hypertension. Pharm J Chin PLA 2013; 29: 384–387.
Google Scholar
69Cao HY, Liu BX, Li LM. Application effect of pharmacy guidance in the senile patients with hypertension and type 2 diabetes. DIABETES NEW WORLD 2017; 20: 6–9.
Google Scholar
70Ma YP, Liu Q. The application of comprehensive nursing intervention on elderly patients with hypertension. Nurs Rehab J 2017; 2: 159–160.
Google Scholar
71Conn VS et al. Interventions to improve medication adherence in hypertensive patients: systematic review and meta-analysis. Curr Hypertens Rep 2015; 17: 94.
10.1007/s11906-015-0606-5
PubMed Web of Science® Google Scholar
72Machado M et al. Sensitivity of patient outcomes to pharmacist interventions. Part II: systematic review and meta-analysis in hypertension management. Ann Pharmacother 2007; 41: 1770–1781.
10.1345/aph.1K311
PubMed Web of Science® Google Scholar
73Morgado MP et al. Pharmacist interventions to enhance blood pressure control and adherence to antihypertensive therapy: review and meta-analysis. Am J Health Syst Pharm 2011; 68: 241–253.
10.2146/ajhp090656
PubMed Web of Science® Google Scholar
74Ruppar TM, Delgado JM, Temple J. Medication adherence interventions for heart failure patients: a meta-analysis. Eur J Cardiovasc Nurs 2015; 14: 395–404.
10.1177/1474515115571213
PubMed Web of Science® Google Scholar
75Ruppar TM et al. Medication adherence interventions among hypertensive black adults: a systematic review and meta-analysis. J Hypertens 2017; 35: 1145–1154.
10.1097/HJH.0000000000001260
PubMed Web of Science® Google Scholar
76Rao Y et al. Current status and future prospects of the development of clinical Pharmacy in China: a SWOT analysis. Pak J Pharm Sci 2016; 29: 415–421.
PubMed Web of Science® Google Scholar
77 National Bureau of Statistics of People's Republic of China. Statistical Yearbook, 2015. http://www.stats.gov.cn/tjsj/ndsj/2016/indexch.htm (accessed 08 October 2017)
Google Scholar

Citing Literature

Volume26, Issue4

August 2018

Pages 291-301

This article also appears in:

Patient Medication Adherence

Filename	Description
ijpp12452-sup-0001-FigS1.docxWord document, 321.3 KB	Figure S1. PRISMA chart of the study selection process.
ijpp12452-sup-0002-TableS1.xlsxMS Excel, 17.7 KB	Table S1. Characteristics of studies included in the meta-analysis.
ijpp12452-sup-0003-AppendixS1.docxWord document, 11.9 KB	Appendix S1. Search strategy (Pubmed).

Interventions to improve medication adherence among Chinese patients with hypertension: a systematic review and meta-analysis of randomized controlled trails